7 I. United States General Accounting Office Washington, D.C Health, Education and Human Services Division

Transcription

1 7 I A0 United States General Accounting Office Washington, D.C Health, Education and Human Services Division B October 2, 1995 The Honorable Max Baucus United States Senate Dear Senator Baucus: One of the issues before the Congress as it considers modifications to the Medicare program is controlling costs while maintaining access to basic hospital and physician services. One model intended to preserve access to basic services in rural areas is medical assistance facilities (MAF), which are limited-service hospitals located only in Montana. Following the closure of numerous rural hospitals in Montana, the state legislature created the MAF provider category in In 1990, the Congress authorized Medicare to pay for MAF services provided to Medicare beneficiaries on the basis of reasonable cost. In 1993, the Congress extended this authorization until July 1, The MAF program was meant to preserve access to basic emergency care, outpatient services, and limited inpatient care in areas where full-service hospitals had closed or were at risk of closure. MAFs must be located in frontier counties1 or be more than 35 miles from another hospital. MAFs were not intended to provide surgical services (although they are not prohibited from doing so), and inpatient stays are limited to 96 hours. Montana established special licensure rules to allow mid-level practitioners (physician assistants and nurse practitioners) to provide care at MAFs under the supervision of a physician, who is not required to be collocated with the MAF. Currently, seven MAFs exist in Montana, primarily in the eastern portion of the state. Each MAF shares space, personnel, and utilities with a nursing home. The MAFs each have an emergency koom, outpatient clinic, and a 2- to lo-bed inpatient unit. 'Under the Montana law, a frontier county is one with a population density of fewer than six persons per square mile. GAO/HEHS-96-12R Montana's Medical Assistance Facilities

2 B You asked us to develop information on cases treated and services performed at MAFs, the relative cost to Medicare for inpatient services at MAFs and at acute-care hospitals, and the number of hospitals that might qualify if the program was expanded nationwide. To address your questions, we obtained Medicare cost and claims data for five of the seven MAFs operating in Montana.' We compared the cost of treating Medicare inpatients3 at MAFs with the cost of treating the same conditions at acute-care hospitals, most of which are paid through Medicare's prospective payment system (PPS); We also obtained Medicare data on outpatient and related professional services performed at MAFs. The cost and claims data covered 172 Medicare inpatient stays and over 8,200 outpatient services at MAFs for various time periods from December 1990 through Details of our analysis and a description of our methodology appear in enclosure 5. We conducted our work between October 1994 and July 1995, in accordance with generally accepted government auditing standards. In summary, MAFs primarily serve patients with urgent, but uncomplicated, conditions or stabilize patients who have more complicated needs before transferring them to fullservice hospitals. Medicare costs for patients served by MAFs were on average lower than if the patients had been treated at regular hospitals. MAFs also serve as primary (and, in some areas, the only) providers of emergency and routine outpatient services for area residents. In 1990, researchers from the University of Minnesota estimated that about 510 hospitals nationwide would meet Montana's qualifying criteria for MAFs. CASES TREATED AND SERVICES FURNISHED BY MAFs All but 1 of the 172 inpatient admissions to MAFs were emergency admissions, which involved treatment for a wide range of medical conditions. The 172 patients were assigned to 67 different medical diagnosis-related groups 'No cost or claims data were available for the two most recently certified MAFs. 3For the period of time for which we obtained data, Medicare patients were 68 percent of all MAF discharges, Medicaid patients were 3 percent, and other categories made up the remaining 29 percent. 2 GAO/HEHS-96-12R Montana's Medical Assistance Facilities

3 B (DRG).' The three medical conditions most commonly treated by the MAFs were pneumonia (22 cases), inflammation of the digestive canal (15 cases), and heart failure and shock (12 cases), which together accounted for 28 percent of the 172 cases. Conditions classified as respiratory, cardiac, digestive, and "diabetes and other metabolic conditions" accounted for 100 of the cases (58 percent) and 30 of the DRGs treated at MAFs. (A summary of inpatient DRGs treated at MAFs appears in enclosure 1.) In addition to providing inpatient care, MAFs provide local primary care for many Medicare beneficiaries. From 1991 to 1994, the 5 MAFs we studied submitted over 8,200 outpatient Medicare claims5 covering more than 23,000 primary care and diagnostic procedures. About 92 percent of the outpatient procedures were for specimen collection for laboratory tests, the laboratory tests themselves, diagnostic - radiology services, and physician services. The MAFs also provided emergency care for injuries such as fractures, open wounds, contusions, strains and sprains, and burns. (The outpatient procedures performed by MAFs from 1991 to 1994 are summarized in enclosure 2.) PAYMENTS TO MAFs LOWER THAN PPS RATES Costs of inpatient care for the 172 Medicare-covered stays compared favorably with the amount Medicare would have paid if those patients had been treated at PPS hospitals. While costs varied among the MAFs, overall costs at the five MAFs were about $75,300 less than the amount Medicare would have paid rural PPS hospitals for treating the same conditions between December 1990 and June As table 1 shows, overall costs for treatment at the five MAFs were about $132,100 less than the amount Medicare would have paid for treating the same conditions at urban hospitals.6 (Enclosure 3 includes more detail on our cost analysis, and 4DRGs are used to classify inpatients into groups that determine the rate of payment under PPS. 51n addition, the physician and physician assistants affiliated with 1 MAF submitted 122 claims for services directly to Medicare, including 35 claims for outpatient services and 87 for hospital visits. 6We used the rate for Billings, Montana, because hospitals in that city received all of the patients who transferred from MAFs to urban acute-care hospitals. 3 GAO/HEHS-96-12R Montana's Medical Assistance Facilities

4 B enclosure 5 describes our methodology for estimating MAF costs and PPS payments.) Table 1: Net Increase [Decrease) in Medicare Costs for 172 MAF Patients Compared With Estimated PPS Payments MAF Dahl Memorial Healthcare Center Garfield County Health Center Net difference of estimated MAF costs compared with PPS payments to rural and urban hospitals Rural hospitals Urban hospitals $(22,200) $(34,008) (16,400) (20,400) McCone County Hospital (64,600) (83,000) Prairie Hospital Roosevelt Hospital Community Memorial 6,000 (5,000) 21,900 10,300 Total net payments $(75,300) I $(132,100) During the period covered by our review, 18 of the 172 inpatients (about 10.5 percent) were transferred from a MAF to an acute-care hospital. During fiscal years 1991 through 1993, about 4.3 percent of all Medicare inpatients at rural hospitals in Montana were transferred to another hospital. We think it is reasonable that the percentage of MAF patients transferred is higher than the percentage transferred from rural hospitals for the following reason: One function of a MAF is to stabilize patients and prepare them for transfer to a facility if treatment beyond the scope of MAF services is needed. An official with the peer review organization for Montana and Wyoming suggested the following two reasons why the transfer rate for MAFs may be higher than that for other rural hospitals: -- patients admitted for observation are transferred after a few hours because their symptoms worsen or they need surgery or -- patients reach the 96-hour limit on length of stay and must be transferred. 4 GAO/HEHS-96-12R Montana's Medical Assistance Facilities

5 B Medicare claims data support these explanations. Of the 18 transfer patients, 10 were transferred within 24 hours of being admitted to the MAF and 4 had been in the MAF for 96 hours. Regardless of what kind of hospital makes the transfer, all transfers result in higher cost to Medicare because two facilities receive payment for the same patient. Under PPS, the transferring hospital receives a per-diem payment determined by dividing the PPS payment by the mean length of stay associated with the patient's DRG. The hospital from which the patient is finally discharged receives the full PPS payment for the patient's DRG. When patients are transferred from MAFs, the MAF receives cost-based reimbursement for the patient, and the hospital from which the patient is finally discharged receives the full PPS payment. We estimate that the costs of treating the 18 transfer cases at the MAFs were about $7,900 greater than the amount Medicare would have paid an acute-care hospital in per diem payments if the patient had first gone to an acute-care hospital for the same length of time. Considering the additional costs associated with transfers reduces our estimate of total MAF savings over PPS payments to about $67,400 when compared with payments to rural hospitals and to about $124,200 when compared with payments to the hospitals in Billings. (Additional information on transfer cases is provided in enclosure 4, and our methodology for estimating costs for transfer patients is described in enclosure 5.) POTENTIAL NUMBER OF MAFs NATIONWIDE Available data suggest that the number of hospitals nationwide that could convert to MAFs is relatively small. Two types of limited service hospitals are currently recognized by Medicare: MAFs in Montana and rural primary care hospitals (RPCH),' which are currently authorized only in California, Colorado, Kansas, New York, North Carolina, 'RPCHs are one provider type under the Essential Access Community Hospital program, which was created by the Omnibus Budget Reconciliation Act of 1989 (P.L , Dec. 19, 1989). A RPCH is limited to six inpatient beds, and inpatient care is limited to an average of 72 hours. 5 GAO/HEHS-96-12R Montana's Medical Assistance Facilities

6 B South Dakota, and West Virginia. In a 1990 study,' researchers associated with the University of Minnesota estimated that a total of about 510 hospitals would meet either the frontier or distance criteria Montana uses as a qualifying condition for MAFs.' The researchers also estimated that, at most, about 370 rural hospitals might convert to a RPCH. But given a variety of factors, the researchers estimated that no more than 100 to 150 rural hospitals across the country would ultimately convert to either a MAF or a RPCH. AGENCY COMMENTS The Health Care Financing Administration reviewed a draft of this letter and provided some technical comments, which we incorporated where appropriate. This letter was prepared under the direction of Thomas Dowdal, Assistant Director. Please contact Mr. Dowdal at (202) or me at (202) if you have any questions. Other analysts who made major contributions to this letter include Robert Sayers, Suzanne Rubins, Roger Hultgren, and Jerry Baugher. Sincerely yours, 'Sarah F. Jaggar Director, Health Financing and Public Health Issues Enclosures - 5 *Jon B. Christianson, Ira S. Moscovice, and Guoyu Tao, Final Report, Medical Assistance Facility Certification Criteria, Division of Health Services Research and Policy, School of Public Health, University of Minnesota (Oct. 1990). 'About 200 hospitals would qualify as MAFs under the distance criterion, about 120 would qualify because they were located in frontier counties, and about 190 would qualify under both criteria. 6 GAO/HEHS-96-12R Montana's Medical Assistance Facilities

7 ENCLOSURE 1 ENCLOSURE 1 SUMMARY OF INPATIENT MAF CASES REVIEWED Cases I DRGs I I DRG category! Number Percent Number Percent Respiratory system conditions Cardiac conditions Digestive system conditions I I Diabetes and other metabolic conditions Other medical conditions Total " " anumbers do not add to total because of rounding. GAO/HEHS-96-12R Montana's Medical Assistance Facilities

8 ENCLOSURE 2 ENCLOSURE 2 SUMMARY OF OUTPATIENT PROCEDURES PERFORMED AT FIVE MAFs, Type of procedure Number of procedure@ Percent Laboratory tests 15, ii Specimen collection for 2, mm laboratory tests.. Diagnostic radiology 1,706 7 I Physician servicesb 1,400 6' Other 1,890 8 I Total 23,125 '1nn ' "The summary of procedures does not include billings by the physician and physician assistants at one MAF, who billed Medicare directly for their services. Between October 1994 and February 1995, they submitted 122 claims, which included 35 claims for outpatient services and 87 claims for hospital visits. bincludes 19 services coded as minor surgical procedures. 8 GAO/HEHS-96-12R Montana s Medical Assistance Facilities

9 ENCLOSURE 3 ENCLOSURE 3 COMPARISON OF MAF COSTS PER DISCHARGE TO PPS- PAYMENTS FOR SIMILAR CASES for rural fmated PPS astimated MAF case was GAO/HEHS-96-12R Montana's Medical Assistance Facilities

10 ENCLOSURE 4 ENCLOSURE 4 SUMMARY OF MAF TRANSFER CASES Acute-care hospital to which MAF Location Number of patients were transferred patients Community Memorial Hospital Sidney, MT 1 Deaconess Hospital Billings, MT 4 Fallon County Medical Complex Baker, MT 1" Glendive Community Hospital Glendive, MT 4 Holy Rosary Hospital Miles City, MT 3 Mercy Hospital Williston, ND 1 St. Vincent's Hospital Billings, MT 3 Trinity Hospital Wolf Point, MT lb Total 18 'This patient was subsequently transferred to Holy Rosary Hospital, Miles City, Montana. *his patient was subsequently transferred to St. Vincent's Hospital, Billings, Montana. 10 GAO/HEHS-96-12R Montana's Medical Assistance Facilities

11 ENCLOSURE 5 ENCLOSURE 5 OBJECTIVES, SCOPE, AND METHODOLOGY Our objectives were to develop information on the cases treated and services performed at medical assistance facilities, the relative cost of providing inpatient health care services to Medicare beneficiaries at MAFs and at acute-care hospitals, and the number of hospitals nationwide that might qualify as limited service facilities if such an option were made available. We did not compare MAF outpatient costs to such costs at other hospitals or to Medicare payments to other types of suppliers, such as clinical laboratories. However, the MAFs are often the only providers of these services in their areas. We obtained automated cost data for five MAFs from three files maintained by the Health Care Financing Administration: -- Hospital Cost Report Information System (HCRIS), which includes selected data from hospital cost reports; -- Medicare Provider Analysis and Review (MEDPAR), for data on the diagnoses and length of stay associated with Medicare patients admitted to MAFs; and -- the standard analytical file, for data on outpatient services provided to Medicare beneficiaries. We estimated the costs for each MAF Medicare inpatient stay. We then compared those costs with the amount Medicare would have paid an acute-care hospital under the prospective payment system for the same diagnosis-related groups at hospitals in rural Montana and the urban hospitals in Billings, Montana. We also compared the amount Medicare would have paid a PPS hospital and a MAF that transferred a patient to an acute-care hospital. ESTIMATING MAF INPATIENT COSTS Because the MAFs in our analysis were certified at different times and had varying cost reporting years, the cost report information we obtained covers different time periods for each facility, as identified in table GAO/HEHS-96-12R Montana's Medical Assistance Facilities

12 ENCLOSURE 5 ENCLOSURE 5 Table 5.1: Number of Cost Reports and Inpatient Claims Data Available for Five MAFs Reviewed MAF Location Certification date Dahl Memorial Ekalaka May 17, 1991 Healthcare Center Garfield County Jordan May 17, 1991 Health Center McCone County Circle Dec. 18, 1990 Hospital Prairie Community Hospital Roosevelt Memorial Hospital Total Terry Jan. 1, 1992 Culbertson Nov. 20, 1992 Number of Number of cost inpatient reports claims in available MEDPAR in HCRIS G We calculated Medicare inpatient operating costs for each MAF's cost reporting period, excluding capital costs.10 We then computed the average daily cost for Medicare patients for each cost reporting period at each facility by dividing operating- costs by the number of Medicare days. We estimated the cost of treating each MAF patient by multiplying the facility's daily Medicare cost by the number of days each patient was an inpatient. l%e excluded capital costs because they are not reimbursed through DRG-based PPS payments. 12 GAO/HEHS-96-12R Montana's Medical Assistance Facilities

13 ENCLOSURE 5 ENCLOSURE 5 ESTIMATING PPS REIMBURSEMENT RATES We calculated the PPS reimbursement rates for the 172 MAF inpatients in our analysis for hospitals located in rural Montana and Billings, Montana.ll We identified each patient's DRG from the MEDPAR file and estimated the amount Medicare would have paid for each of the 172 MAF discharges in a rural and urban PPS hospital, using PPS payment rates in effect during fiscal years 1991 through Our estimate of PPS payments does not include adjustments for teaching status or disproportionate share of low-income patients, either or both of which a particular hospital might receive. INPATIENTS WHO TRANSFERRED FROM MAFs TO PPS HOSPITALS Eighteen patients were treated at a MAF then transferred to a PPS hospital. We estimated Medicare's cost of treating those patients at the MAF as we did for all patients, that is, by multiplying the MAF's daily Medicare cost by the number of days the patient was at the MAF prior to transfer. PPS hospitals are reimbursed for the care provided to a patient who transfers to another hospital according to a per-diem rate. This rate is obtained by dividing the PPS payment by the mean length of stay expected for the patient's DRG (this number is published annually with the DRG relative weights). We calculated the per-diem PPS rate for each of the 18 transfer cases and multiplied that amount by the number of days each patient stayed at the MAF prior to transfer. The result of this calculation was the estimated payment that PPS hospitals would have received had the patient been treated at a PPS hospital for the same number of days that the patient was at the MAF. Our estimate of the cost of treating the patients at the MAF before transferring them was the difference between the estimated cost of the case at the MAF and the estimated cost of treating the patient at a PPS hospital before transfer. (106425) llpps hospitals in Billings are paid urban rates. 13 GAO/HEHS-96-12R Montana's Medical Assistance Facilities

14

15 Ordering Information The first copy of each GAO report and testimony is free. Additional copies are $2 each. Orders should be sent to the following address, accompanied by a check or money order made out to the Superintendent of Documents, when necessary. VISA and Mastercard credit cards are accepted, also. Orders for 100 or more copies to be mailed to a single address are discounted 25 percent. Orders by mail: U.S. General Accounting Office P.O. Box 6015 Gaithersburg, MD or visit: Room th St. NW (corner of 4th and G Sts. NW) U.S. General Accounting Office Washington, DC Orders may also be placed by caliing (202) or by using fax number (301) , or TDD (301) Each day, GAO issues a list of newly available reports and testimony. To receive facsimile copies of the daily list or any list from the past 30 days, please caii (202) using a touchtone phone. A recorded menu wili provide information on how to obtain these Iists. For information on how to access GAO reports on the INTERNET, send an message with info in the body to: info@ PRINTED RECYCLED PAPER

16 United States General Accounting Office Washington, D.C Official Business Penale for Private Use $300 Address Correction Requested